Multi-phase adaptive network configuration
View Patent ↗The present invention enables fast reconfiguration of paths associated with bufferless core nodes in a network where any two edge nodes may interconnect through a core node. In order to enhance network agility and performance, edge nodes associated with a core node are separated into a plurality of groups based on, for example, round-trip delay from the core node. This separates the short-haul paths from the long haul paths, thereby enabling the short-haul paths to be configured more frequently than the long-haul paths. It is preferable that the periods be set such that each additional period is an integer multiple of the previous one. This allows for simultaneous configuration of groups of paths. Although the methods and apparatus of the present invention are most effective for bufferless optical switches, they are still effective for electronic switches which may be bufferless or equipped with input data buffers.
1. A method of adaptively configuring a network, said network having at least one core node and a plurality of edge nodes, the method comprising the steps of:
time-locking each of said edge nodes to each of said at least one core node;
grouping edge nodes of the plurality of edge nodes according to a reconfiguration latency of each edge node, to produce a plurality of edge-node groups;
assigning a group reconfiguration-latency threshold to each edge node group; and
periodically reconfiguring paths associated with edge nodes in said each edge node group at said group reconfiguration-latency threshold.
2. A method according to claim 1 wherein said edge-node groups are ordered according to said group reconfiguration-latency threshold and wherein the group reconfiguration-latency threshold of each succeeding edge-node group is selected to be an integer multiple of the group reconfiguration-latency threshold of a preceding edge-node group.
3. A method according to claim 1 wherein each edge node of said plurality of edge nodes further comprises a source node and wherein a the reconfiguration latency of said each edge node of said plurality of edge nodes comprises a round trip delay between said each edge node of said plurality of edge nodes and said each of said at least one core node.
4. A method according to claim 1 wherein said plurality of edge nodes further comprises source nodes and sink nodes and wherein the reconfiguration latency of each edge node of said plurality of edge nodes comprises a round trip delay between said each edge node of said plurality of edge nodes and a one of said sink nodes over paths traversing said each of said at least one core node.
5. A method according to claim 1 wherein said group reconfiguration-latency threshold for each of said edge-node groups is dynamically derived from network measurements.
6. A method according to claim 5 wherein said network measurements comprise round-trip latencies.
7. An apparatus for adaptively configuring a network, said network having at least one core node and a plurality of edge nodes, said apparatus comprising:
means for time-locking each of said edge nodes to each of said at least one core node;
means for grouping edge nodes of the plurality of edge nodes, according to a reconfiguration latency of each edge node, to provide a plurality of edge node groups;
means for assigning a group reconfiguration-latency threshold to each edge node group; and
means for periodically reconfiguring paths associated with edge nodes in said each edge node group at said group reconfiguration-latency threshold.
8. An apparatus according to claim 7 wherein the edge-node groups are ordered according to said group reconfiguration-latency threshold and wherein the group reconfiguration-latency threshold of each succeeding edge-node group is selected to be an integer multiple of the group reconfiguration-latency threshold of a preceding edge-node group.
9. An apparatus according to claim 7 wherein each edge node of said plurality of edge nodes further comprises a source node and wherein a the reconfiguration latency of said each edge node of said plurality of edge nodes includes a round trip delay between said each edge node of said plurality of edge nodes and said each of said at least one core node.
10. An apparatus according to claim 7 wherein said plurality of edge nodes further comprises source nodes and sink nodes and wherein a the-reconfiguration latency of each edge node of said plurality of edge nodes comprises a round trip delay between said each edge node of said plurality of edge nodes and a one of said sink nodes over paths traversing said each of said at least one core node.
11. An apparatus according to claim 7 wherein said group reconfiguration-latency threshold for each of said edge-node groups is dynamically derived based on network measurements.
12. An apparatus according to claim 7 wherein said each of said at least one core node is an optical core node.
13. An apparatus according to claim 7 wherein said each of said at least one core node is an electronic core node.
14. An apparatus according to claim 13 wherein said electronic core node is provided with a data buffer.
15. An apparatus according to claim 12 wherein said means for time-locking each of said edge nodes to each of said at least one core node comprises means for time locking at least one wavelength channel in a wavelength division multiplexing path between one of said edge nodes and said optical core node.